During my photo safari last year to Rocky Mountain National Park, we had one of those scenes where you just aren’t sure what to do. The clouds were rolling in, and the light turned flat and gloomy. Yet we could see the light hitting Long’s Peak in the distance, and it looked really neat. I shot a bunch of images but none of them looked very inspiring. In fact, they were flat and boring, just like the light. Continue reading Photo of the Day: Rocky Mountain Sunset
I’ve always been drawn to Nikon’s “performance” DSLRs. With the exception of my D2x, I’ve preferred models that deliver high ISO performance and frame rates over sheer pixel count. As part of my continuing exploration of the new Nikon D4, I’ll take a look at how it performs as a landscape camera, especially in light of the new D800, which clearly is getting lots of well-deserved attention these days.
When people think of good landscape cameras, they usually focus on a couple of features:
- Megapixels (resolution)
- Dynamic Range (ability to capture tones)
- Base ISO (lower ISO can be more flexible for creative slow exposures)
To compare sensors, let’s take a look at the DXO Mark comparisons of the Nikon D800, D4, and D2Xs:
On specs alone, it’s pretty clear that the new Nikon D800 comes out as a clear winner here (other than base ISO of 100) against the D4, as it can deliver a stunning 13.23 EV of dynamic range at ISO 100! The D4 is certainly no slouch at 12.58 EV, as it bests the D2Xs by nearly two full stops. Consider for a moment that many photographers, myself included, considered the D2X a fabulous landscape camera.
The chart above also shows another truth about DSLRs: their DR performance drops off as you boost ISO sensitivity. In the case of the D2Xs, the drop-off is quite pronounced. The sharp decline in DR seen in the D2Xs is something you can also see in the D800. Notice, however, that the D4 sensor does not show the immediate drop-off in DR performance that the other two cameras do. In fact, between ISO 100 and ISO 1600 on the D4, you only give up 1.29 EV of dynamic range. The D800 loses 3.2 EV of DR over the same range.
The D800 delivers incredible dynamic range, but only when you shoot at ISO 100.
So what? I thought landscape photographers always use the camera’s base ISO, with a tripod, etc. Well, this is true for many scenes, there are other situations when you start to use higher ISOs, like when you have subject movement that you don’t want blurred. What the chart above tells me is that while the D800 kicks the D4 at ISO 100, once I’m at ISO 200 the playing field is level, and at ISO 400+, the D4 will have more DR. In fact, the D4 has nearly as much DR at ISO 3200 as the D2Xs did at ISO 100. That is called flexibility.
I’m not even going to argue this point. If you want pixels, the D800 wins handily. There is no doubt that the resolution of the D800 is amazing, and for detailed landscape work, you might even consider the D800e without the anti-aliasing filter. I thought long and hard about this feature as I pre-ordered my D4, as I certainly do my fair share of landscape images on workshops, etc. Then I thought about how I present and share the majority of my images: online. Yes, I have an Epson 7800 printer with a 24″ paper path. I love it. But when I print, I print at 16×24″ or smaller most of the time, with most of my prints being in the 12×18″ range. And frankly, I’ve been very happy with the print quality at 12MP at these sizes. I won’t lie; I don’t make my living making gallery prints. So that is something that is a personal preference to me. When most people see your images on-screen at 1080 pixels or less, does having 36-MP matter?
The other great option with the D800 is cropping flexibility. Needless to say, you can get a 15MP image from a DX-crop using the D800. But I’ll say it again: cropping is no substitute for proper technique. Sure, there are times when having the flexibility to crop is nice, but if I was using a 36-MP camera to make 15MP images, that would seem like a waste to me.
The only major downside of a high-resolution camera for what I do is file size. Every time you click the shutter of the D800, you generate a 41MB RAW file (14-bit, lossless compressed). This large file size limits the D800 to 4fps shooting, which isn’t bad, but is still slower than the 10fps afforded by the D4. Is that important?
Traditional landscape photographers don’t care about frame rate. “Give me 3fps and that’s more than I’ll ever use for landscapes,” they say. True, for single-shot landscapes frame rate is irrelevant. But what about for HDR bracketing? When you merge images to create HDR, as I often do, it’s critical that there be as little difference between frames as possible. Otherwise, you’ll get motion artifacts. So for that kind of shooting, it’s ideal to have a fast shutter speed (to prevent motion blur) and a fast frame rate. In that situation, the D4 can deliver 14-bit NEFs at 10fps, and do so with faster shutter speeds (assuming you’re using a higher ISO).
Speaking of HDR, one nice feature I stumbled upon is that the D4 will do 2 and 3-EV increments for exposure bracketing, while the D800 is still (I don’t know why) limited to 1-EV spacing. For many HDR scenes, all you really need are three frames: -2,0,+2EV. It always annoyed me using the D3 that I needed five frames (1EV apart) to get a three-frame series. When you combine this feature with 10fps shooting, you can further minimize potential movement artifacts between shots in the bracketed series.
Time to produce a ±2EV bracketed sequence (t=0 at first frame):
- D4 (3 shots, 2EV apart, 10fps): 0.2s
- D3s (5 shots, 1EV apart, 9fps): 0.444s
- D800 (5 shots, 1EV apart, 4fps): 1.0s
With the D4, I could theoretically capture four complete ±2EV bracketed sequences in the time it takes the D800 to capture a single sequence. Of course, this isn’t what I’m trying to do when bracketing; the important thing is that the likelihood of movement artifacts is reduced when you only need to capture three frames in 0.2s.
The Nikon D4 and D800 are both fantastic technological achievements, and each serves a fundamentally different niche. For traditional landscape photographers, I can think of no better camera than the D800/D800e, provided that it’s used at base ISO most of the time. For shooters who explore different types of creative outlets, like HDR, shooting hand-held, or just want the extra degrees of freedom for controlling shutter speed, then the Nikon D4 holds its own in just about all aspects except sheer resolution. Of course, at twice the price, it may be difficult to justify a D4 over the D800 as a primary landscape camera. I do lots of non-landscape shooting (sports, wildlife, indoor portraits) that makes the D4 a great choice for my own needs; you should evaluate your own needs before buying any camera.
[youtube_sc url=”http://youtu.be/Mi_klUkJByw” title=”Photographer%20Jason%20P.%20Odell%20explains%20how%20autofocus%20influences%20the%20Nikon%203D%20Matrix%20metering%20system%20in%20digital%20cameras” border=”1″ autohide=”1″ hd=”1″]
Ever wonder why you can get different exposures on the exact same scene with your Nikon camera using 3D Matrix Metering? The meter and AF system are linked in a way that usually produces great results. But for landscape photographers, it can sometimes cause overexposure if you focus on dark parts of the scene, like a shady foreground.
Nikon’s 3D matrix metering is an advanced exposure system that evaluates the entire color image. The same scene can be exposed differently, depending on the location of the active AF point. When the AF point falls on a dark part of the scene, the meter tends to open up the exposure (brighten it). If the AF point is on a bright part of the scene, the camera will expose the scene slightly darker. This difference in exposure can be particularly important to landscape photographers, who usually choose a foreground object for the focus point. If the foreground is relatively dark, the camera will often blow out the sky.
The solution is to first focus on the dark object in the foreground, and then lock focus (I use the AF-On button technique for this). Next, move the active AF point to a bright part of the scene. This allows the meter to bias the exposure to preserve highlight details in the final image.
Periodically, I like to go through my old images and re-process them; it’s one of the great benefits of shooting in RAW. Being able to work with my old images in new software really opens up some options that I didn’t even consider at the time I made the shot. I captured this image in 2005 while on a photo safari in South Dakota. At the time, I was shooting a new Nikon D2x camera and I had just gotten my 17-55mm f/2.8 AFS DX zoom Nikkor lens. I also had just upgraded to Photoshop CS2, which had a new feature: “Merge to HDR.” I thought HDR could be a cool thing to learn, so I shot a lot of bracketed exposure sequences during this 5-day trip. Many of them were uninspiring. Others, I found difficult if not impossible to process, and so I just processed the best exposure in the sequence with traditional techniques (that’s the nice thing about bracketing– you’ll always have at least one “normal” exposure).
The one thing I didn’t do, however, was delete the other exposures from the bracketed sequence. They’ve just been sitting on one of my 1TB hard drives, waiting for me to give them a second chance. Fast forward to 2011, and now HDR tools have progressed to the point where you can get great results quickly and easily. So, yesterday I went back and re-processed the HDR sequence in HDR Efex Pro and Capture NX 2, and I was quite pleased! Here’s what was in the new technology that I couldn’t get in 2005. Continue reading Don’t Stop Thinking About Tomorrow